Aprimoramento do manejo avícola utilizando rede neural artificial: um estudo de caso real

Abstract

This work shows how machine learning techniques can be used to improve poultry management activities. Poultry management consists of the control of resources such as water, feed, temperature, humidity, ventilation, among others. These resources are managed by a human specialist known as a poultry farmer. The way these resources are controlled influences the development of birds, impacting productivity. On-the-spot surveys show that poultry farmers with more expertise often achieve better productivity results. Therefore, a learning model using linear regression through an artificial neural network is used to assist poultry farmers in the decision-making process. The result is a specialist system that suggests to the poultry farmer the resources to be used.